The molecular mechanisms by which eukaryotes regulate gene expression and chromatin structure are important for basic scientific knowledge that is necessary for understanding many complex biological phenomena including development and human disease. This proposal addresses two related issues of broad significance: the molecular mechanisms of transcriptional repression by the evolutionarily related yeast Cyc8-Tup1 and human TLE corepressors, and the relationship between global patterns of histone modifications to transcription and maintenance of epigenetic states. To do this, we will combine chromatin immunoprecipitation, genetic, genomic, and biochemical approaches to carry out the following projects. First, we will address the mechanisms and relative importance of three general models for how Cyc8-Tup1 mediates repression: interaction with the Mediator complex that associates with RNA polymerase II; direct effects on chromatin; steric interference. We will determine the mechanistic step(s) inhibited by Cyc8-Tup1, analyze whether Cyc8-Tup1 is sufficient to recruit Mediator and cooperates with or inhibits activators from recruiting Mediator, identify direct targets of the Tup1 repression domain within the Mediator complex, and address whether Cyc8-Tup1 recruits histone deacetylases and how it mediates long-range effects on chromatin. Second, we will test the model that Eaf3 controls the global pattern of histone acetylation in yeast cells via an interaction between Eaf3 chromodomains and methylated histones in the coding region that causes preferential deacetylation by the Rpd3 complex. We will also test the idea that an unappreciated level of transcription throughout the yeast genome accounts for the apparent paradox that the Pall transcriptional elongation complex is required for genome-wide methylation of histone H3 at lysines 4 and 79. Third, we will test whether histone deposition during S phase occurs by the conventional model involving H3-H4 tetramers or a newly proposed model involving H3-H4 dimers, an issue with major implications for how epigenetic states are maintained through cell division. In addition, we will directly address the relationship and mechanism between transcriptional activity and histone dissociation, and measure the rate of transcription-associated (i.e. non-replicative) nucleosome deposition. Fourth, as part of our continuing collaboration with Tom Gingeras at Affymetrix that involves tiled microarrays, we will define, in an unbiased manner and on a whole-genome scale, the physiological targets of human TLE proteins in HL60 cells undergoing retinoic acid-induced differentiation. These results will be linked to detailed maps of RNA transcripts, histone modifications, chromatin-modifying activities, and transcriptional regulatory factors that will be performed on the identical samples in work not covered by this proposal. We will use RNAi to identify genes affected by the various TLE proteins, and will address the mechanism of repression by TLE proteins by using derivatives of the well characterized B-interferon promoter.